U.S. patent application number 12/627196 was filed with the patent office on 2010-06-03 for adjusting bids based on predicted performance.
This patent application is currently assigned to Google Inc.. Invention is credited to Scott S. Benson, Adrian Dumitru Corduneanu, James A. Gallagher, Tomas Lloret Llinares, Karl Robert Pfleger, Abhinay Sharma, Andrew E. Silverman.
Application Number | 20100138291 12/627196 |
Document ID | / |
Family ID | 42223667 |
Filed Date | 2010-06-03 |
United States Patent
Application |
20100138291 |
Kind Code |
A1 |
Silverman; Andrew E. ; et
al. |
June 3, 2010 |
Adjusting Bids Based on Predicted Performance
Abstract
A predicted performance metric of a candidate advertisement in
an advertising slot can be received and compared to a baseline
predicted performance metric for the candidate advertisement. A
target bid associated with the candidate advertisement can be
adjusted based upon the comparison, and the adjusted bid can be
submitted to an auction for the advertising slot.
Inventors: |
Silverman; Andrew E.; (San
Francisco, CA) ; Sharma; Abhinay; (Mountain View,
CA) ; Gallagher; James A.; (Mountain View, CA)
; Llinares; Tomas Lloret; (Mountain View, CA) ;
Benson; Scott S.; (Santa Clara, CA) ; Pfleger; Karl
Robert; (San Francisco, CA) ; Corduneanu; Adrian
Dumitru; (Mississauga, CA) |
Correspondence
Address: |
FISH & RICHARDSON P.C.
PO BOX 1022
MINNEAPOLIS
MN
55440-1022
US
|
Assignee: |
Google Inc.
Mountain View
CA
|
Family ID: |
42223667 |
Appl. No.: |
12/627196 |
Filed: |
November 30, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61119192 |
Dec 2, 2008 |
|
|
|
Current U.S.
Class: |
705/14.45 ;
705/14.41; 705/14.46; 705/14.71 |
Current CPC
Class: |
G06Q 30/0242 20130101;
G06Q 30/0275 20130101; G06Q 30/0247 20130101; G06Q 30/0246
20130101; G06Q 30/02 20130101 |
Class at
Publication: |
705/14.45 ;
705/14.71; 705/14.41; 705/14.46 |
International
Class: |
G06Q 30/00 20060101
G06Q030/00 |
Claims
1. A computer implemented method, comprising: receiving a predicted
performance metric for a candidate advertisement in an advertising
slot; comparing the predicted performance metric for the candidate
advertisement in the advertising slot to a baseline predicted
performance metric of the candidate advertisement; adjusting a
target bid associated with the candidate advertisement based upon
the comparison of the predicted performance metric for the
candidate advertisement and the baseline predicted performance
metric for the candidate advertisement; and submitting the adjusted
target bid to an auction for the advertising slot.
2. The computer implemented method of claim 1, wherein the
predicted performance metric is based upon historical conversion
tracking data for previous advertising slots.
3. The computer implemented method of claim 2, wherein the
predicted performance metric is further based upon historical
conversion tracking data for the current advertisement in past
advertising slots similar to the current advertising slot.
4. The computer implemented method of claim 1, further comprising
identifying the baseline predicted performance metric.
5. The computer implemented method of claim 4, wherein identifying
the baseline predicted performance metric comprises: receiving
predicted conversion rates for a set of previous auctions
associated with the candidate advertisement; and determining an
average predicted conversion rate from the set of previous auctions
and associated predicted conversion rates; wherein the baseline
predicted performance metric comprises the average predicted
conversion rate.
6. The computer implemented method of claim 5, further comprising
identifying those previous auctions in which the candidate
advertisement received a click as the set of previous auctions for
which predicted conversion rates are received.
7. The computer implemented method of claim 5, further comprising
identifying those previous auctions in which the candidate
advertisement received a conversion as the set of previous auctions
for which predicted conversion rates are received.
8. The computer implemented method of claim 5, further comprising
identifying those previous auctions in which the candidate
advertisement was served as the set of previous auctions for which
predicted conversion rates are received.
9. The computer implemented method of claim 5, further comprising
identifying those previous auctions in which the candidate
advertisement was also a candidate advertisement as the set of
previous auctions for which predicted conversion rates are
received.
10. The computer implemented method of claim 1, wherein: comparing
the predicted performance metric of the candidate advertisement to
the baseline predicted performance metric for the candidate
advertisement comprises dividing the predicted performance metric
by the baseline predicted performance metric to derive an
adjustment ratio; and adjusting the target bid comprises
multiplying the target bid by the adjustment ratio.
11. The computer implemented method of claim 1, further comprising:
determining an average price paid for placement of the
advertisement; comparing the average price paid to the target bid;
if the average price paid exceeds the target bid, throttling the
adjusted bid.
12. The computer implemented method of claim 11, wherein throttling
the adjusted bid comprises decreasing the adjusted bid to the
target bid.
13. The computer implemented method of claim 11, wherein throttling
the adjusted bid comprises decreasing the adjusted bid to a maximum
adjusted bid.
14. The computer implemented method of claim 11, wherein throttling
the adjusted bid comprises discounting the adjusted bid by a
percentage based on how much the average price paid exceeds the
target bid.
15. A computer-readable medium having instructions stored thereon,
which, when executed by a processor, causes the processor to
perform operations comprising: receiving a predicted conversion
rate for a candidate advertisement in the current advertising slot;
comparing the predicted conversion rate for the advertisement in
the current advertising slot to a baseline predicted conversion
rate; adjusting the target bid associated with the advertisement
based upon the comparison of the predicted conversion rate and the
baseline predicted conversion rate; and submitting the adjusted
target bid to the impending auction.
16. The computer-readable medium of claim 15, wherein the predicted
conversion rate is based upon historical conversion tracking data
for the current advertisement in past advertising slots similar to
the current advertising slot.
17. The computer-readable medium of claim 15, further comprising
identifying the baseline predicted conversion rate.
18. The computer-readable medium of claim 17, wherein identifying
the baseline predicted conversion rate comprises: identifying a set
of previous auctions; receiving predicted conversion rates for the
set of previous auctions; and determining an average predicted
conversion rate from the set of previous auctions and associated
predicted conversion rates; wherein the baseline predicted
conversion rate comprises the average predicted conversion
rate.
19. The computer-readable medium of claim 15, wherein comparing the
predicted conversion rate to the baseline rate comprises dividing
the predicted conversion rate by the baseline predicted conversion
rate to derive an adjustment ratio, and wherein adjusting the
target bid comprises multiplying the target bid by the adjustment
ratio.
20. The computer-readable medium of claim 15, further comprises:
determining an average discount bid associated with the
advertisement, wherein the average discount bid is the average
amount paid by the advertiser for the advertisement; comparing the
average discount bid to the target bid; and if the average discount
bid exceeds the target bid, throttling the adjusted bid.
21. A computer implemented method, comprising: receiving a
predicted click-through rate for a candidate advertisement in an
available advertising slot; comparing the predicted click-through
rate for the candidate advertisement in the available advertising
slot to an average predicted click-through rate for the candidate
advertisement derived from a control set; and adjusting a target
impression based bid associated with the candidate advertisement
based upon the comparison of the predicted click-through rate for
the candidate advertisement and the average predicted click-through
rate for the candidate advertisement; submitting the adjusted
target impression based bid to an auction for the available
advertising slot.
22. A computer implemented method, comprising: receiving a
predicted revenue metric for a candidate advertisement in an
available advertising slot; comparing the predicted revenue metric
for the advertisement in the available advertising slot to an
average predicted revenue metric for the candidate advertisement
derived from a control set; and adjusting a target conversion based
bid associated with the candidate advertisement based upon the
comparison of the predicted revenue metric for the candidate
advertisement and the average predicted revenue metric for the
candidate advertisement; submitting the adjusted target bid to an
auction for the available advertising slot.
Description
CROSS-REFERENCED TO RELATED APPLICATIONS
[0001] This application claims the benefit under 35 U.S.C.
.sctn.119(e) of U.S. Patent Application No. 61/119,192, titled
"ADJUSTING BIDS BASED ON PREDICTED PERFORMANCE" filed Dec. 2, 2008,
which is incorporated herein by reference.
TECHNICAL FIELD
[0002] The subject matter of this application is generally related
to online advertising.
BACKGROUND
[0003] The rise of the Internet has enabled access to a wide
variety of content items, e.g., video and/or audio files, web pages
for particular subjects, news articles, etc. Such access to these
content items has likewise enabled opportunities for targeted
advertising. For example, content items of particular interest to a
user can be identified by a search engine in response to a user
query. The query can include one or more search terms, and the
search engine can identify and, optionally, rank the content items
based on the search terms in the query and present the content
items to the user (e.g., according to the rank). This query can
also be an indicator of the type of information of interest to the
user. By comparing the user query to a list of keywords specified
by an advertiser, it is possible to provide targeted advertisements
to the user.
[0004] Another form of online advertising is advertisement
syndication, which allows advertisers to extend their marketing
reach by distributing advertisements to additional partners. For
example, third party online publishers can place an advertiser's
text or image advertisements on web pages that have content related
to the advertisement. Because the users are likely interested in
the particular content on the publisher webpage, they are also
likely to be interested in the product or service featured in the
advertisement. Accordingly, such targeted advertisement placement
can help drive online customers to the advertiser's website.
[0005] Advertisers can bid for placements based upon how much the
advertiser values the placement. In some examples, the advertiser
can bid based upon impressions of the advertisement. In such
examples, the advertiser is charged whenever the advertisement is
served. In other examples, the advertise can bid based upon a
click-through for the advertisement. In such examples, the
advertiser is charged only when a user clicks on the advertisement
after the advertisement is served to the user.
SUMMARY
[0006] In general, the subject matter of this application relates
to adjusting target bids for advertisement in an auction. Methods
for target bid adjustment can include: identifying a candidate
advertisement, the candidate advertisement being a candidate for
participating in an auction of a current advertising slot, the
candidate advertisement being associated with an advertiser and a
target bid; receiving a predicted performance metric for the
advertisement in the current advertising slot; comparing the
predicted performance metric for the advertisement in the current
advertising slot to a baseline predicted performance metric of the
candidate advertisement; adjusting the target bid associated with
the advertisement based upon the comparison of the predicted
performance metric and the baseline predicted performance metric;
and submitting the adjusted target bid to the auction. Other
embodiments of this aspect include corresponding methods,
apparatus, and computer program products.
DESCRIPTION OF DRAWINGS
[0007] FIG. 1 is a block diagram of an implementation of an online
advertising environment.
[0008] FIG. 2 is a block diagram of an example implementation of an
advertisement server shown in FIG. 1.
[0009] FIG. 3 is a flow diagram of an implementation of a bid
adjustment process.
[0010] FIG. 4 is block diagram of an example computer system.
[0011] Like reference numbers and designations in the various
drawings indicate like elements.
DETAILED DESCRIPTION
[0012] Bids for advertisements can be adjusted based upon a
predicted performance of the advertisement in a current
advertisement placement being auctioned relative to a baseline
predicted performance of the advertisement over a set of past
auctions or placements. The predicted performance, for example, can
include a predicted conversion rate, a predicted click-through
rate, or a predicted revenue or profit margin, among many others.
The set of past auctions or placements, for example, can be
selected based upon other auctions in which the advertisement was a
candidate for participation in the auction, placements in which the
advertisement received an impression, placements in which the
advertisement received a click, or placements in which the
advertisement received a conversion, among many others. The bid can
be adjusted up or down depending upon the comparison between the
predicted performance and the baseline predicted performance. The
adjusted bid can be submitted to the impending auction for the
current advertisement placement.
Advertising System Overview
[0013] FIG. 1 is a block diagram of an implementation of an online
advertising system 100. In some implementations, one or more
advertisers 102 can directly, or indirectly, enter, maintain, and
track advertisement ("ad") information in an advertising management
system 104. The ads may be in the form of graphical ads, such as
banner ads, text only ads, image ads, audio ads, video ads, ads
combining one of more of any of such components, etc. The ads may
also include embedded information, such as links, meta-information,
and/or machine executable instructions. One or more publishers 106
may submit requests for ads to the system 104. The system 104
responds by sending ads to the requesting publisher 106 for
placement on one or more of the publisher's web properties (e.g.,
websites and other network-distributed content).
[0014] Other entities, such as users 108 and the advertisers 102,
can provide usage information to the system 104, such as, for
example, whether or not a conversion or click-through related to an
ad has occurred. This usage information can include measured or
observed user behavior related to ads that have been served. The
system 104 performs financial transactions, such as crediting the
publishers 106 and charging the advertisers 102 based on the usage
information.
[0015] A computer network 110, such as a local area network (LAN),
wide area network (WAN), the Internet, or a combination thereof,
connects the advertisers 102, the system 104, the publishers 106,
and the users 108.
[0016] One example of a publisher 106 is a general content server
that receives requests for content (e.g., articles, discussion
threads, music, video, graphics, search results, web page listings,
information feeds, etc.), and retrieves the requested content in
response to the request. The content server may submit a request
for ads to an ad server in the system 104. The ad request may
include a number of ads desired. The ad request may also include
content request information. This information can include the
content itself (e.g., page or other content document), a category
corresponding to the content or the content request (e.g., arts,
business, computers, arts-movies, arts-music, etc.), part or all of
the content request, content age, content type (e.g., text,
graphics, video, audio, mixed media, etc.), geo-location
information, etc.
[0017] In some implementations, the content server can combine the
requested content with one or more of the ads provided by the
system 104. This combined content and ads can be sent to the user
108 that requested the content for presentation in a viewer (e.g.,
a browser or other content display system). The content server can
transmit information about the ads back to the ad server, including
information describing how, when, and/or where the ads are to be
rendered (e.g., in HTML or JavaScript.TM.).
[0018] Another example publisher 106 is a search service. A search
service can receive queries for search results. In response, the
search service can retrieve relevant search results from an index
of documents (e.g., from an index of web pages). Search results can
include, for example, lists of web page titles, snippets of text
extracted from those web pages, and hypertext links to those web
pages, and may be grouped into a predetermined number of (e.g.,
ten) search results.
[0019] The search service can submit a request for ads to the
system 104. The request may include a number of ads desired. This
number may depend on the search results, the amount of screen or
page space occupied by the search results, the size and shape of
the ads, etc. In some implementations, the number of desired ads
will be from one to ten, or from three to five. The request for ads
may also include the query (as entered or parsed), information
based on the query (such as geo-location information, whether the
query came from an affiliate and an identifier of such an
affiliate), and/or information associated with, or based on, the
search results. Such information may include, for example,
identifiers related to the search results (e.g., document
identifiers or "docIDs"), scores related to the search results
(e.g., information retrieval ("IR") scores), snippets of text
extracted from identified documents (e.g., web pages), full text of
identified documents, feature vectors of identified documents, etc.
In some implementations, IR scores can be computed from, for
example, dot products of feature vectors corresponding to a query
and a document, page rank scores, and/or combinations of IR scores
and page rank scores, etc.
[0020] The search service can combine the search results with one
or more of the ads provided by the system 104. This combined
information can then forwarded to the user 108 that requested the
content. The search results can be maintained as distinct from the
ads, so as not to confuse the user between paid advertisements and
presumably neutral search results.
[0021] Finally, the search service can transmit information about
the ad and when, where, and/or how the ad was to be rendered back
to the system 104.
[0022] As can be appreciated from the foregoing, the advertising
management system 104 can serve publishers 106, such as content
servers and search services. The system 104 permits serving of ads
targeted to documents served by content servers. For example, a
network or inter-network may include an ad server serving targeted
ads in response to requests from a search service with ad spots for
sale. Suppose that the inter-network is the World Wide Web. The
search service crawls much or all of the content. Some of this
content will include ad spots (also referred to as "inventory")
available. More specifically, one or more content servers may
include one or more documents. Documents may include web pages,
email, content, embedded information (e.g., embedded media),
meta-information and machine executable instructions, and ad spots
available. The ads inserted into ad spots in a document can vary
each time the document is served or, alternatively, can have a
static association with a given document.
Target Bid Adjustment
[0023] In some implementations, the system 104 includes an
adjustable bidding system which predicts or estimates a performance
metric for a given ad placement for an advertisement using
performance prediction engine. In further implementations, the
system 104 includes a bid adjustment system used to adjust bid
associated with an advertisement and specified by the advertiser
102 to account for the potential that a current advertising slot
provides a better opportunity for the advertiser 102 than other
opportunities.
[0024] The advertising management system includes an ad server 120
operable to serve advertisements to publishers 106 through a search
engine interface 130 or through an ads for content interface 140.
Advertisements for the search engine interface 130 can be
identified, for example, based upon the search terms entered by the
user 108. Advertisements for the ads for content interface 140 can
be identified, for example, based upon an analysis of the content
of the publisher 106 web page and a comparison to keywords or
concepts associated with the advertisements.
[0025] In some implementations, the advertising management system
104 can include an advertisement data store 150 and a campaign data
store 160. Advertisements and associated usage data can be stored
as advertisement data in an advertisement data store 150. In some
implementations, an advertiser 102 can further manage the serving
of advertisement by specifying an advertising campaign. The
advertising campaign can be stored in campaign data in a campaign
data store 160 that can, for example, specify advertising budgets
for advertisements, when, where and under what conditions
particular advertisements may be served for presentation, etc. For
example, a computer company may design an advertising campaign for
a new laptop computer that is scheduled to be released in several
weeks. The advertising campaign may have a budget of $500,000, and
may have 30 different advertisements that are to be served for
presentation during the month of November. Such data defining the
advertisement campaign can be stored in the campaign data 160.
[0026] In some implementations, the advertisement server 120 can be
coupled to a performance tracking data store 170. The advertisement
server 120 can send performance tracking data identifying how well
the advertisement is performing to the performance tracking data
store 170. Identification of how well the ad is performing can be
based on a performance metric. The performance metric can be
defined, for example, in terms of the number of times the
advertisement has been served, the number of times the
advertisement has been selected (e.g., a click-through rate), the
number of conversions that have been recorded, the average profit
or revenue associated with an impression or selection of the
advertisement, survey information about the advertisement from
those viewers that have received the advertisement, or predicted
performance scores for each participation in an auction, among
others.
[0027] A performance prediction system 180 can use the performance
tracking data from the performance tracking data store 170 to
predict performance of an advertisement in a given slot. The
predicted performance information, for example, can be used by the
ad server 120 to adjust a bid associated with an advertisement.
[0028] In some implementations, the ad server 120 can adjust the
bid associated with an advertisement based upon comparing a
predicted performance metric for the advertisement with a baseline
predicted performance metric for the advertisement. In various
examples, the predicted performance metric can be any of a
selection of the advertisement (e.g., a click-through), a
conversion (e.g., sale) associated with an impression, or revenue
or profit generated as a result of the impression, among many
others.
[0029] In some implementations, the baseline predicted performance
metric can be identified by identifying a control set. The control
set, for example, can be a set of previous action data for a set of
auctions for which the advertisement was a candidate advertisement,
a set of auctions for which the advertisement received an
impression, a set of impressions which resulted in the
advertisement being selected (e.g., clicked) by a user, a set of
impressions which resulted in a conversion, etc. Upon identifying
the control set, the ad server 120 can average predicted
performance metrics for each of the auctions in the control set to
identify the baseline predicted performance. For example, if the
predicted performance metric is predicted conversion rate (pCVR),
the baseline predicted performance metric can be identified as:
Baseline predicted performance = pCVR 1 + pCVR 2 + pCVR N N
##EQU00001##
Where pCVR.sub.n corresponds to the predicted conversion rate for
the nth member of the control set, and N corresponds to the
cardinality of the control set. Thus, for example, if the control
set covers three auctions {A1, A2, A3}, the predicted performance
metric is predicted conversion rate, and the pCVR for A1 is 30%,
the pCVR for A2 is 5%, and the pCVR for A3 is 10%, the baseline
predicted performance metric can be identified as 15%
((pCVR.sub.A1+pCVR.sub.A2+pCVR.sub.A3)/3=(30%+5%+10%)/3=15%). In
those examples using a different predicted performance metric, the
predicted conversion rate can be replaced by the appropriate
predicted performance metric.
[0030] The ad server 120 can compare the predicted performance
metric of a candidate advertisement for a current advertisement
placement associated with an impending auction to the baseline
predicted performance metric. Based upon the comparison, the ad
server 120 can adjust the bid associated with the advertisement. In
some implementations, the comparison can include dividing the
predicted performance metric for the current placement by the
baseline predicted performance metric. In such implementations, any
error introduced by a performance prediction component is canceled
out because the same error can be assumed to be present in both the
predicted performance metric for advertisement in the current
advertising slot and the baseline predicted performance metric for
the advertisement. The result of dividing the predicted performance
metric by the baseline predicted performance metric can identify an
adjustment ratio. In those implementations using predicted
conversion rates as the predicted performance metric, the
adjustment ratio can be identified by the equation:
Adjustment Ratio = pCVR current pCVR baseline ##EQU00002##
Where pCVR.sub.current is the predicted conversion rate for the
current advertising slot, and pCVR.sub.baseline is the an average
predicted conversion rate for the advertisement over a control set.
For example, if predicted conversion rate for a current advertising
slot is 15% and a baseline predicted conversion rate for the
current advertising slot is 10%, the adjustment ratio can be
identified as 15%/10%, or 1.5. The adjustment ratio can be used to
adjust the bid associated with the candidate advertisement. In some
implementations, the bid can be multiplied by the adjustment ratio
to derive an adjusted bid, which can be submitted to the impending
auction.
[0031] In some implementations, the baseline predicted performance
metric can be continuously updated. In such implementations, the ad
server 120 can serve a portion of the requests (e.g., 5%, 15%, 25%,
50%, etc.) for which an advertisement qualifies as a candidate
advertisement without adjusting the bid. The portion of the
requests which are served without adjusting the bid can be
identified as the control set.
[0032] In some implementations, the portion of requests allocated
to the control set can be dynamically adjusted. For example, when
an advertisement is new and there does not exist enough tracking
data to provide an adequate sampling for the control set, the ad
server 120 can allocate an increased percentage of traffic (e.g.,
50%) to the control set. After a predetermined period of time, for
example, or number of impressions (or other metric), the percentage
of traffic allocated to the control set can be reduced by the ad
server 120. In another example, if an advertisement does not
accumulate enough tracking data, the percentage of traffic
allocated to the control group can remain high.
[0033] In some implementations, the control set can be replaced by
a sliding window identifying a specified number of previous
auctions preceding the current auction. Thus, a specified number of
previous auctions can be used to update the baseline predicted
performance metric.
[0034] In some implementations, if an average price paid by the
advertiser 102 exceeds the target bid specified by the advertiser
102, the ad server 120 can throttle the adjustment of the target
bid. In other implementations, if the average price paid by the
advertiser 102 exceeds the target bid by a threshold amount, the ad
server 120 can throttle the adjustment of the target bid.
Throttling the adjustment of the target bid can reduce the average
price paid by the advertiser 102 for a placement, click-through,
conversion, etc. to the target bid.
[0035] In some implementations, throttling of the adjustment to the
target bid can be done based upon an amount by which the average
price paid exceeds the target bid. For example, if an average price
paid exceeds a target bid by 20%, the adjusted bid can be
discounted by 20%. Thus, for example, if a target bid specified by
the advertiser is $1.00 on a CPC basis, and an adjustment ratio is
identified as 1.2, the adjusted bid of $1.20 can be discounted by
20% to $0.96 (e.g., Throttled Bid=Adjusted Bid*(1-Discount
Percentage)=$1.20*(1-0.20)=$0.96).
[0036] In other implementations, throttling of the adjusted target
bid can be done by identifying a maximum adjusted bid above which
the target bid cannot be adjusted. In some examples, the maximum
adjusted bid can be set based upon the target bid. For example, the
target bid may be raised by a percentage (e.g., 20%) to identify
the maximum adjusted bid. In other examples, the maximum adjusted
bid may be the target bid itself.
[0037] In other implementations, throttling of the adjusted bid can
be accomplished by allocating a larger portion of the requests for
the candidate advertisement to the control set. Allocation of a
larger portion of the bids to the control set can provide a maximum
bid as the target bid for a larger percentage of the auctions for
which the advertisement is a candidate, and thereby lowering the
average price paid for the advertisement.
[0038] FIG. 2 is a block diagram of an example implementation of an
ad server 120 shown in FIG. 1. In some implementations, the ad
server 120 can include an ad server interface 200, a candidate
advertisement identification component 210, a bid adjustment
component 220 and an auction component 230. The ad server interface
200 can receive advertisement requests from one or more of the
search engine interface 130 or the ads for content interface 140.
Receipt of an advertisement request signals an impending auction
for a current advertisement slot on the requesting web page or
search engine page. The impending auction can be based upon any
auction paradigms including, for example, a cost per impression
metric (e.g., cost per mille (CPM) or cost per action (e.g., cost
per click (CPC), cost per conversion, etc.) paradigm. The auction
paradigm can be based upon publisher preference or by an ad
exchange on which the ad slot is auctioned.
[0039] Upon receiving a request for an advertisement, the candidate
advertisement identification component 210 can identify a candidate
advertisement for the request. The candidate advertisement can be
identified, for example, based upon search terms included in a
search query or based upon content of a requesting web page. In
some implementations, candidate advertisements can also be
identified based upon preferences of publishers for types of
advertisements or preferences of advertisers for types of slots.
The candidate advertisement identification component 210 can
retrieve advertisement information from the ad data store 150
and/or campaign information from the campaign data store 160 in
order to facilitate identification of candidate advertisements.
[0040] For those candidate advertisements identified by the
candidate advertisement identification component 210 that have
enabled adjustment of an associated target bid, a bid adjustment
component 220 can adjust the target bid based upon information from
the performance prediction system 180. In some implementations,
enablement of target bid adjustment can be based upon an advertiser
enabling performance tracking. In other implementations, enablement
of target bid adjustment can be based upon explicit instruction
from the advertiser to enable target bid adjustment (e.g., based
upon input received through an advertiser interface to the ad
server 120).
[0041] In some implementations, bid adjustment can be based upon
comparison of a predicted performance metric (e.g., predicted
click-through rate, predicted conversion rate, predicted revenue,
predicted profit margin, etc.) to a baseline predicted performance
metric. For example, in a cost per click auction model, the
predicted conversion rate for the advertisement in the slot being
auctioned can be compared to a baseline predicted conversion rate
for the advertisement across a sampling of auctions. The sampling
of auctions can be identified, for example, by allocating a portion
of auctions for which the advertisement is a candidate
advertisement to a control set. Alternatively, the control set can
be limited to those auctions where the candidate advertisement
registered an impression, a click-through, a conversion, or any
other measurable action. The baseline predicted performance metric
can be an average of the predicted performance metrics for each of
the auctions/placements identified by the control set.
[0042] In some implementations, the comparison of the predicted
performance metric for the current advertisement placement with the
baseline predicted performance metric can include dividing the
predicted performance metric for the current advertisement
placement by the baseline predicted performance metric. For
example, if the average predicted conversion rate for the candidate
advertisement in the auctions/placements in the control group is
10% and the predicted conversion rate for the candidate
advertisement in the impending auction/current placement is 5%, an
adjustment ratio of 1/2 can be identified (e.g., 5%/10%). A bid
associated with the candidate advertisement can thereby be scaled
by the adjustment ratio. Thus, if the target bid for the candidate
advertisement is $1.00 on CPC basis, the adjusted bid is the target
bid multiplied by the adjustment ratio, or $0.50 on a CPC basis
(e.g., $1.00.times.1/2).
[0043] In some implementations, the bid adjustment component 220
can throttle the adjusted bid based upon an average price paid
exceeding the target bid or an average price paid exceeding the
target bid by a threshold margin. For example, if the target bid is
$1.00, and the average price paid is $1.35, future adjusted bids
can be modified to reduce the average price paid toward the target
bid. In some implementations, throttling of the adjusted bid can
reduce the average price paid while maintaining a bias for the
advertisement towards placements in which the bid is expected to be
more successful (e.g., based upon the comparison the predicted
performance metric to the baseline predicted performance
metric).
Bid Adjustment Process
[0044] FIG. 3 is a flow diagram of an implementation of a bid
adjustment process. At stage 310 candidate advertisements can be
identified for the auction. Candidate advertisements can be
identified, for example, by a candidate advertisement
identification component (e.g., candidate advertisement
identification component 210 of FIG. 2). In some implementations,
candidate advertisements can be identified based upon search terms
received from a search engine. In other implementations, candidate
advertisements can be identified based upon analysis of the content
of a web page requesting the advertisement. The analysis of the web
page can occur in real-time or offline when the publisher engages
an ad server to supply future advertisements for the publisher. In
examples where the analysis of a web page occurs offline, the ad
server can supply a snippet to the publisher for insertion into the
web page. The snippets upon execution by a browser can cause the
browser to send a request for an advertisement along with an
identification, identifying the web page. The ad server can use the
identification to retrieve the previous analysis of the web page.
Candidate advertisements can then be selected based upon the
retrieved web page analysis. In some implementations, candidate
advertisements can be further identified based preferences of the
publisher and/or advertiser.
[0045] At stage 320, a predicted performance is received. The
predicted performance can be received, for example, by a bid
adjustment component (e.g., bid adjustment component 220 of FIG. 2)
from a performance prediction system (e.g., performance prediction
system 180 of FIGS. 1 and 2). The performance of the advertisement
is predicted with respect to the current placement (e.g., the
placement for which the ad server received an advertisement
request). In some implementations, the predicted performance of the
advertisement can be based upon past performance of the
advertisement in similar slots (e.g., on the same web page, for the
same search query, at the same time of day, for similar users, for
similar content or concepts, etc.).
[0046] At stage 330, the received predicted performance is compared
to a baseline predicted performance. The received predicted
performance can be compared to the baseline predicted performance,
for example, by a bid adjustment component (e.g., bid adjustment
component 220 of FIG. 2). In some implementations, the baseline
predicted performance can be compiled from a control set of
placements for which the same advertisement was a candidate. The
control set of placements, in some examples, can be identified as a
portion of placements for which the advertisement is a candidate
advertisement which do not adjust the target bid. In various
examples, the control set can be limited to those placements that
did not use an adjusted target bid and which registered a
click-through, conversion or other activity with respect to the
advertisement. In some implementations, the comparison of the
predicted performance to the baseline predicted performance can
include dividing the predicted performance by the baseline
predicted performance to derive an adjustment ratio. Other forms of
comparison and or adjustment can be used, including, for example,
logarithmic functions. Logarithmic functions can provide an
asymptotic relationship between the adjusted bid and the target
bid, thereby inhibiting the adjustment of the target bid beyond the
asymptote specified by the function.
[0047] At stage 340, a target bid can be adjusted. The target bid
can be adjusted, for example, by a bid adjustment component (e.g.,
bid adjustment component 220 of FIG. 2). In those implementations
that derive an adjustment ratio, the adjustment ratio can be used
to adjust the target bid specified by the advertiser. For example,
the adjustment ratio can be multiplied by the target bid to
identify an adjusted bid. In some implementations, when an average
price paid by the advertiser for placements of the advertisement
exceeds the target bid or is at a threshold above the target bid,
the adjusted bid can be throttled to reduce the average price paid
by the advertiser for placement of the advertisement. In such
implementations, the bid adjustment system can continue to bias the
placements in favor of more favorable placements by applying a
discount to all adjusted bids, rather than only those adjusted bids
which exceed the target bid, or some other mechanism.
[0048] At stage 350, the adjusted bid is submitted to an auction.
The adjusted bid can be submitted to an auction, for example, by a
bid adjustment component (e.g., bid adjustment component 220 of
FIG. 2) in conjunction with an auction component (e.g., auction
component 230 of FIG. 2). In some implementations, the auction
component can determine whether or not the advertisement qualifies
for placement based upon the adjusted bid. If an advertisement
qualifies for placement, the advertisement can be communicated to
the publisher or search engine through an ad server interface.
[0049] In some implementations, the predicted performance metric
can be replaced by past performance. For example, if tracking data
indicates that an advertisement, on average, has a conversion rate
of 5%, the baseline predicted performance metric can be identified
as 5%. Similarly, if the tracking data indicates that the
advertisement has previously had an 8% conversion rate in this
particular slot (or in similar slots on similar properties), an
adjustment ratio of 1.6 can be derived.
Example Computer System
[0050] FIG. 4 is block diagram of an example computer system 500.
The system 500 includes a processor 510, a memory 520, a storage
device 530, and an input/output device 540. Each of the components
510, 520, 530, and 540 can, for example, be interconnected using a
system bus 1350. The processor 510 is capable of processing
instructions for execution within the system 500. In one
implementation, the processor 510 is a single-threaded processor.
In another implementation, the processor 510 is a multi-threaded
processor. The processor 510 is capable of processing instructions
stored in the memory 520 or on the storage device 530.
[0051] The memory 520 stores information within the system 500. In
one implementation, the memory 520 is a computer-readable medium.
In one implementation, the memory 520 is a volatile memory unit. In
another implementation, the memory 520 is a non-volatile memory
unit.
[0052] The storage device 530 is capable of providing mass storage
for the system 500. In one implementation, the storage device 530
is a computer-readable medium. In various different
implementations, the storage device 530 can, for example, include a
hard disk device, an optical disk device, or some other large
capacity storage device.
[0053] The input/output device 540 provides input/output operations
for the system 500. In one implementation, the input/output device
540 can include one or more of a network interface devices, e.g.,
an Ethernet card, a serial communication device, e.g., and RS-232
port, and/or a wireless interface device, e.g., and 802.11 card. In
another implementation, the input/output device can include driver
devices configured to receive input data and send output data to
other input/output devices, e.g., keyboard, printer and display
devices 560. Other implementations, however, can also be used, such
as mobile computing devices, mobile communication devices, set-top
box television client devices, etc.
[0054] The advertising management system 104 and/or ad server 120,
and components thereof, can be realized by instructions that upon
execution cause one or more processing devices to carry out the
processes and functions described above. Such instructions can, for
example, comprise interpreted instructions, such as script
instructions, e.g., JavaScript or ECMAScript instructions, or
executable code, or other instructions stored in a computer
readable medium. The advertising management system 104 and/or ad
server 120, and components thereof, can be distributively
implemented over a network, such as a server farm, or can be
implemented in a single computer device.
[0055] Embodiments of the subject matter and the functional
operations described in this specification can be implemented in
digital electronic circuitry, or in computer software, firmware, or
hardware, including the structures disclosed in this specification
and their structural equivalents, or in combinations of one or more
of them. Embodiments of the subject matter described in this
specification can be implemented as one or more computer program
products, i.e., one or more modules of computer program
instructions encoded on a tangible program carrier for execution
by, or to control the operation of, data processing apparatus.
[0056] A computer program (also known as a program, software,
software application, script, or code) can be written in any form
of programming language, including compiled or interpreted
languages, or declarative or procedural languages, and it can be
deployed in any form, including as a stand alone program or as a
module, component, subroutine, or other unit suitable for use in a
computing environment. A computer program does not necessarily
correspond to a file in a file system. A program can be stored in a
portion of a file that holds other programs or data (e.g., one or
more scripts stored in a markup language document), in a single
file dedicated to the program in question, or in multiple
coordinated files (e.g., files that store one or more modules, sub
programs, or portions of code). A computer program can be deployed
to be executed on one computer or on multiple computers that are
located at one site or distributed across multiple sites and
interconnected by a communication network.
[0057] Additionally, the logic flows and structure block diagrams
described in this patent document, which describe particular
methods and/or corresponding acts in support of steps and
corresponding functions in support of disclosed structural means,
may also be utilized to implement corresponding software structures
and algorithms, and equivalents thereof. The processes and logic
flows described in this specification can be performed by one or
more programmable processors executing one or more computer
programs to perform functions by operating on input data and
generating output.
[0058] Processors suitable for the execution of a computer program
include, by way of example, both general and special purpose
microprocessors, and any one or more processors of any kind of
digital computer. Generally, a processor will receive instructions
and data from a read only memory or a random access memory or both.
The essential elements of a computer are a processor for performing
instructions and one or more memory devices for storing
instructions and data. Generally, a computer will also include, or
be operatively coupled to receive data from or transfer data to, or
both, one or more mass storage devices for storing data, e.g.,
magnetic, magneto optical disks, or optical disks. However, a
computer need not have such devices.
[0059] Computer readable media suitable for storing computer
program instructions and data include all forms of non volatile
memory, media and memory devices, including by way of example
semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory
devices; magnetic disks, e.g., internal hard disks or removable
disks; magneto optical disks; and CD ROM and DVD ROM disks. The
processor and the memory can be supplemented by, or incorporated
in, special purpose logic circuitry.
[0060] To provide for interaction with a user, embodiments of the
subject matter described in this specification can be implemented
on a computer having a display device, e.g., a CRT (cathode ray
tube) or LCD (liquid crystal display) monitor, for displaying
information to the user and a keyboard and a pointing device, e.g.,
a mouse or a trackball, by which the user can provide input to the
computer. Other kinds of devices can be used to provide for
interaction with a user as well; for example, feedback provided to
the user can be any form of sensory feedback, e.g., visual
feedback, auditory feedback, or tactile feedback; and input from
the user can be received in any form, including acoustic, speech,
or tactile input.
[0061] Embodiments of the subject matter described in this
specification can be implemented in a computing system that
includes a back end component, e.g., as a data server, or that
includes a middleware component, e.g., an application server, or
that includes a front end component, e.g., a client computer having
a graphical user interface or a Web browser through which a user
can interact with an implementation of the subject matter described
is this specification, or any combination of one or more such back
end, middleware, or front end components. The components of the
system can be interconnected by any form or medium of digital data
communication, e.g., a communication network. Examples of
communication networks include a local area network ("LAN") and a
wide area network ("WAN"), e.g., the Internet.
[0062] The computing system can include clients and servers. A
client and server are generally remote from each other and
typically interact through a communication network. The
relationship of client and server arises by virtue of computer
programs running on the respective computers and having a client
server relationship to each other.
[0063] While this specification contains many specific
implementation details, these should not be construed as
limitations on the scope of any invention or of what may be
claimed, but rather as descriptions of features that may be
specific to particular embodiments of particular inventions.
Certain features that are described in this specification in the
context of separate embodiments can also be implemented in
combination in a single embodiment. Conversely, various features
that are described in the context of a single embodiment can also
be implemented in multiple embodiments separately or in any
suitable subcombination. Moreover, although features may be
described above as acting in certain combinations and even
initially claimed as such, one or more features from a claimed
combination can in some cases be excised from the combination, and
the claimed combination may be directed to a subcombination or
variation of a subcombination.
[0064] Similarly, while operations are depicted in the drawings in
a particular order, this should not be understood as requiring that
such operations be performed in the particular order shown or in
sequential order, or that all illustrated operations be performed,
to achieve desirable results. In certain circumstances,
multitasking and parallel processing may be advantageous. Moreover,
the separation of various system components in the embodiments
described above should not be understood as requiring such
separation in all embodiments, and it should be understood that the
described program components and systems can generally be
integrated together in a single software product or packaged into
multiple software products.
[0065] Particular embodiments of the subject matter described in
this specification have been described. Other embodiments are
within the scope of the following claims. For example, the actions
recited in the claims can be performed in a different order and
still achieve desirable results. As one example, the processes
depicted in the accompanying figures do not necessarily require the
particular order shown, or sequential order, to achieve desirable
results. In certain implementations, multitasking and parallel
processing may be advantageous.
[0066] This written description sets forth the best mode of the
invention and provides examples to describe the invention and to
enable a person of ordinary skill in the art to make and use the
invention. This written description does not limit the invention to
the precise terms set forth. Thus, while the invention has been
described in detail with reference to the examples set forth above,
those of ordinary skill in the art may effect alterations,
modifications and variations to the examples without departing from
the scope of the invention.
* * * * *